Structured network pruning is a practical approach to reduce computation cost directly while retaining the CNNs' generalization performance in real applications. However, identifying redundant filters is a core problem in structured network pruning, and current redundancy criteria only focus on individual filters' attributes. When pruning sparsity increases, these redundancy criteria are not effective or efficient enough. Since the filter-wise interaction also contributes to the CNN's prediction accuracy, we integrate the filter-wise interaction into the redundancy criterion. In our criterion, we introduce the filter importance and filter utilization strength to reflect the decision ability of individual and multiple filters. Utilizing this new redundancy criterion, we propose a structured network pruning approach SNPFI (Structured Network Pruning by measuring Filter-wise Interaction). During the pruning, the SNPFI can automatically assign the proper sparsity based on the filter utilization strength and eliminate the useless filters by filter importance. After the pruning, the SNPFI can recover pruned model's performance effectively without iterative training by minimizing the interaction difference. We empirically demonstrate the effectiveness of the SNPFI with several commonly used CNN models, including AlexNet, MobileNetv1, and ResNet-50, on various image classification datasets, including MNIST, CIFAR-10, and ImageNet. For all experimental CNN models, nearly 60% of computation is reduced in a network compression while the classification accuracy remains.

翻译：结构化网络剪枝是一种实用方法，可直接减少实际应用中的计算成本，同时保持CNN的泛化性能。然而，识别冗余滤波器是结构化网络剪枝的核心问题，现有冗余准则仅关注单个滤波器的属性。随着剪枝稀疏性增加，这些冗余准则有效性或效率不足。由于滤波器间交互同样影响CNN的预测精度，我们将滤波器间交互整合到冗余准则中。在该准则中，我们引入滤波器重要性和滤波器利用强度，分别反映单个及多个滤波器的决策能力。基于这一新冗余准则，我们提出结构化网络剪枝方法SNPFI（基于滤波器间交互测量的结构化网络剪枝）。在剪枝过程中，SNPFI能依据滤波器利用强度自动分配适当稀疏度，并通过滤波器重要性消除无用滤波器。剪枝后，SNPFI通过最小化交互差异，无需迭代训练即可有效恢复剪枝模型的性能。我们在多种常用CNN模型（包括AlexNet、MobileNetv1和ResNet-50）及图像分类数据集（包括MNIST、CIFAR-10和ImageNet）上实证了SNPFI的有效性。对于所有实验CNN模型，在网络压缩中约60%的计算量被削减，同时分类精度保持不变。